Functions to manipulate dates and count days for quantitative finance analysis. The ‘quantdates’ package considers leap, holidays and business days for relevant calendars in a financial context to simplify quantitative finance calculations, consistent with International Swaps and Derivatives Association (ISDA) regulations.
To install the latest development builds directly from GitHub, run this instead:
if (!require("devtools"))
install.packages("devtools")
devtools::install_github("quantilma/quantdates")
# AddDate allows Date input as character object
AddDate(date = '2019-10-04',addDays=14,addMonths=2,addYears=3)
# Add business days for Bogota (CO)
AddBusinessDays(date = Sys.Date(),numDate = 15,loc = 'BOG')
# Return last day of the month for the date
LastDayOfMonth(date = '2020-02-03')
# Returns business days within given range for the location and Dates as character
BusinessDays(loc='BOG', from='2020-10-10', to='2020-11-10')
# Transform numeric dates from Excel
NumExcel2DateR(as.numeric(Sys.Date()))
# Transform numeric dates from R
NumR2DateR(as.numeric(Sys.Date()))
# Number of days between two dates considering leap days.
difftime_leap_year(tfinal='2023-03-05',tinitial='2019-02-28',leapDatesIn=TRUE)
# Number of days between two dates skipping leap days.
difftime_leap_year(tfinal='2023-03-05',tinitial='2019-02-28',leapDatesIn=FALSE)
# Number of business days between two dates in Bogota (CO).
difftime_business(tfinal='2023-03-08',tinitial='2019-02-28',wd=wdBOG)
# Financial day count between two dates according to ACT/365 convention
day_count(tfinal='2023-03-08',tinitial='2019-02-28',convention='ACT/365')
Currently we are developing the guidelines for contributions.
The quantdates package as a whole is licensed under the GPLv3.